Reconocimiento visual del patrón de articulación de palabras en español: un estudio conductual

Several evidences seem to indicate that, during oral comprehension, speech processing is performed on the basis of auditory verbal information and visuo-gestural information coming from the interlocutor. In this framework, the time and accuracy of responses of hearing subjects were evaluated in a lip-reading task with the acoustic pattern suppressed. In this task, 25 adult participants, without a diagnosis of linguistic or cognitive pathologies or hearing or visual impairment, had to distinguish whether they were facing a Spanish word based on the visualization of its articulation. For the task, 240 stimuli were constructed, divided into three groups of 80 Spanish ‘Words’, 80 Spanish ‘Pseudo-words’ and 80 ‘Non-words’, presented randomly. The results regarding response time indicated that the ‘Words’ (m=717.6ms) were recognized at a slower speed than the ‘Pseudowords’ (m=809.5ms) and the ‘Non-Words’ (7633.7ms); while the ‘Pseudowords’ took longer to be responded to than the ‘Non-Words’. On the other hand, in the accuracy results, the stimuli presented the following recognition pattern: ‘Words’<’Pseudowords’<’Non-Words’ (corresponding to 59.89%, 62.5% and 86.3%, respectively). Similarly, robust evidence was found for a relationship between the type of stimulus presented and the hit/miss of responses, although the model indicates that the ‘Non-Word’ condition is more easily recognized and that the difference between ‘Words’ and ‘Pseudowords’ is not relevant. Likewise, a relationship was observed between lower speed and response inaccuracy. These data corroborate the hypothesis of the integration of information from visual perception and language processing, at the same time allowing future work to be projected with electrophysiological evidence.